codebase-research

Pass

Audited by Gen Agent Trust Hub on Apr 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is purely instructional and contains no executable scripts or commands that would pose a security risk.
  • [SAFE]: No patterns of data exfiltration or unauthorized network access were found; the skill only advises searching the local codebase.
  • [SAFE]: No attempts to bypass AI safety guidelines or perform prompt injection were identified.
  • [SAFE]: The use of external tools like github-search is suggested in a standard, non-malicious context for pattern discovery.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 5, 2026, 07:45 PM